# nlpyang/PreSumm

code for EMNLP 2019 paper Text Summarization with Pretrained Encoders

Repository: https://github.com/nlpyang/PreSumm
Canonical: https://ross.abutalabs.com/products/presumm
Language: Python
License: MIT
License Family: permissive
Last push: 2024-07-25T10:16:21+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2575, "days_push": 769, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1303, forks 452 (observed 2026-08-28T04:04:18.300785+00:00)

## What it is
PreSumm is the official PyTorch implementation of the EMNLP 2019 paper 'Text Summarization with Pretrained Encoders', providing BERT-based extractive (BertSumExt) and abstractive (BertSumAbs, BertSumExtAbs) summarization models. It includes training, evaluation, and inference code with pretrained checkpoints for CNN/DailyMail and XSum.

## Use cases
- summarize long documents with bert
- train an extractive summarization model
- run abstractive text summarization with pretrained encoders
- reproduce emnlp 2019 summarization results on cnn/dailymail
- summarize raw text files from the command line
- evaluate summarization models with rouge

## When to choose
- you need BERT-based extractive or abstractive summarization with published ROUGE benchmarks
- you want to reproduce or build on the PreSumm paper's models
- you have GPU resources and are comfortable with older PyTorch versions

## When to avoid
- you need a production-ready, actively maintained summarization library
- you require modern PyTorch or transformer library versions
- you want plug-and-play summarization without training or setup effort

## Facets
- artifact type: library
- maturity: maintenance
- function: nlp, machine-learning, deep-learning
- domain: machine-learning, deep-learning
- platform: python
- tags: text-summarization, bert, abstractive-summarization, extractive-summarization, research-code, pytorch, emnlp-2019, rouge, natural-language-processing, linux, gpu

## Member repositories
- nlpyang/PreSumm (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:18.300785+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:51:13.310798+00:00, confidence not recorded.
  - readme: https://github.com/nlpyang/PreSumm (fetched 2026-08-28T04:04:18.300785+00:00, sha 91cbbf3f1ab6)
- Data as of 2026-08-30T08:39:29.467469+00:00.